Dermacentor Reticulatus in the UK

This electronic thesis or dissertation has been downloaded from Explore Bristol Research, http://research-information.bristol.ac.uk

Author: Bryer, Katie E Title: Effects of temperature and humidity on the mortality of the tick Dermacentor reticulatus in the UK

General rights Access to the thesis is subject to the Creative Commons Attribution - NonCommercial-No Derivatives 4.0 International Public License. A copy of this may be found at https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode This license sets out your rights and the restrictions that apply to your access to the thesis so it is important you read this before proceeding. Take down policy Some pages of this thesis may have been removed for copyright restrictions prior to having it been deposited in Explore Bristol Research. However, if you have discovered material within the thesis that you consider to be unlawful e.g. breaches of copyright (either yours or that of a third party) or any other law, including but not limited to those relating to patent, trademark, confidentiality, data protection, obscenity, defamation, libel, then please contact [email protected] and include the following information in your message:

•Your contact details •Bibliographic details for the item, including a URL •An outline nature of the complaint

Your claim will be investigated and, where appropriate, the item in question will be removed from public view as soon as possible.

Effects of temperature and humidity on the mortality of

the tick Dermacentor reticulatus in the UK

Katherine E. Bryer

A dissertation submitted to the University of Bristol in accordance with the requirements of the degree of MSc (Res) in the Faculty of Life Sciences, School of

Biological Sciences.

August 2020 Word Count: 13,073

Abstract

The disease canine babesiosis, transmitted by the tick Dermacentor reticulatus, has recently been found for the first time in the UK. This disease outbreak highlighted our lack of knowledge of the distribution and abundance of this tick species in the UK, which is believed to be expanding under the influence of climate change. However, given our limited understanding of the influence of microclimatic variables on its activity, it is difficult to predict its future spread to manage any potential risks. The research presented in this thesis aimed to quantify the effects of temperature and relative humidity on the mortality of D. reticulatus. Adult ticks were collected between September 2019 and February 2020 from 4 locations in the UK: West Wales, Essex, North Devon, and South Devon. Different concentrations of potassium hydroxide were used to create a range of humidities inside sealed desiccator jars, each one kept in incubators at temperatures of 4°C, 15°C or 30°C. Five relative humidities were tested: 20%, 40%, 60%, 80% and 95%. Each desiccator contained three tubes of five ticks which were checked every 2 or 3 days for mortality over a period of 9 weeks (63 days). Mortality increased significantly over time. The data show that tick survival rate was lower when the temperature was higher, the humidity was lower, and the longer the ticks had been exposed to these conditions in the desiccator. Integrating temperature and humidity to calculate a measure of saturation deficit showed that there was no effect on tick survival rate until day 20 and that where the saturation deficit value was lower, indicating a high air moisture content, the survival rate increased. This study suggests that D. reticulatus is able to survive at relatively low temperatures, lower than those preferred by other species of ticks such as I. ricinus and I. hexagonus, and that an environmental humidity exceeding 80% RH is required for D. reticulatus ticks to prevent water loss via evaporation. These results have implications for predictions about how D. reticulatus populations and tick-borne disease will be affected by climate change in the UK. It may be that D. reticulatus will spread northwards in the UK seeking cooler temperatures, perhaps more rapidly than other UK species such as I. hexagonus and I. ricinus which tolerate warmer environments. However, this will also be strongly affected by changes in precipitation. Higher rainfall will allow D. reticulatus survival even at elevated average temperature. As a result, pathogens transmitted by D. reticulatus will become more abundant and more widely established in areas of the UK they are not usually found in. Higher temperatures may also cause more successful incubation of pathogens with D. reticulatus, leading to a higher risk of disease.

Acknowledgments

There are several people I would like to thank who have been invaluable in the production of this thesis. Firstly, I wish to express my gratitude to my supervisor Professor Richard Wall for his support, encouragement, and the time he readily gave me throughout the year. I would also like to give special thanks to Bryony Sands for her continuous valuable advice and guidance, especially with R, and without whom, field trips for tick collection would not have been half as much fun. Within the Life Sciences Department, I would like to thank my friends Amber-Rose Cooper, Maddie Noll and many others for the many weeks of encouragement, support and laughs along the way. I am also grateful to Amber for assistance with tick collection in Wales. Lastly, I would like to thank my parents, friends and flatmates for their support and assistance throughout the year.

Author’s Declaration

I declare that the work in this dissertation was carried out in accordance with the requirements of the University’s Regulations and Code of Practice for Research Degree Programmes and that it has not been submitted for any other academic award. Except where indicated by specific reference in the text, the work is the candidate’s own work. Work done in collaboration with, or with the assistance of, others, is indicated as such. Any views expressed in the dissertation are those of the author.

SIGNED……………………...... DATE………………………………

Table of Contents CHAPTER 1- Introduction: Ticks and tick-borne disease 1.1 Systematics of ticks ...... 8 1.2 The Ixodida: morphology and life cycle ...... 9 1.3 How ticks transmit disease ...... 11 1.4 Ticks in the UK ...... 12 1.5 Dermacentor reticulatus ...... 14 1.5.1 Morphology and life cycle ...... 14 1.5.2 Distribution ...... 16 1.5.3 Epidemiological impact ...... 19 1.5.4 Risks to veterinary and public health ...... 20 1.5.5 Environmental requirements ...... 22 1.5.6 Changing abundance and distribution ...... 27 1.6 Study aims ...... 28

CHAPTER 2- Effects of temperature and humidity on mortality 2.1 Introduction ...... 30 2.2 Methods ...... 30 2.2.1 Collection sites ...... 30 2.2.2 Collection method ...... 32 2.2.3 Experiment design ...... 33 2.2.4 Statistical analysis ...... 36 2.3 Results ...... 36 2.4 Discussion ...... 47

CHAPTER 3- Discussion 3 Discussion ...... 51 References ...... 56

List of Figures

CHAPTER 1- Introduction: Ticks and tick-borne disease

1.1 The three-host life cycle of female Ixodidae ticks ...... 9 1.2 Dermacentor reticulatus (A) female (B) male ...... 14 1.3 Current D. reticulatus geographical distribution in Europe detailing where D. reticulatus is present, has been introduced, and is absent based on data provided by the Vector-Net project and published historical data...... 17 1.4 Current known distribution of Dermacentor reticulatus in the United Kingdom using data from Public Health England, the Big Tick Project and field collections between 2009-2016. The symbols show where Dermacentor reticulatus populations were reported to be by the mentioned sources ...... 18

CHAPTER 2- Effects of temperature and humidity on mortality 2.1 Collection of ticks: a) a male D. reticulatus on the blanket, having attached while the blanket was being dragged over vegetation; b) the blanket used for tick collection, made up of a 1.2m bamboo pole and 1m2 cotton sheet; c) D. reticulatus, a male and a female, within a collection tube ready for transportation to the laboratory ...... 33 2.2 Desiccator jar assemblage: a) 5 ticks and a filter paper ‘grass stem’ within a tube; b) fine mesh to allow air flow was secured on top of a tube with a rubber-band to hold it in place; c) two tubes within a desiccator jar standing on top of mesh to prevent them falling into the solution below; d) a desiccator jar containing tubes of ticks with the lid sealed on top using Vaseline to keep the internal humidity from changing ...... 35 2.3 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 10. The error bars represent standard error of the mean...... 37 2.4 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 20 (30°C: Y=0.0787*X - 2.645, Z14=4.6, P<0.001, r²=0.7479). The error bars represent standard error of the mean...... 38 2.5 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 30(30°C: Y=0.0619*X - 2.4079, Z14=4.2, P<0.001, r²=0.7261; 15°C: Y= 0.0538*X - 1.2394, Z14=3.7, P<0.001, r²= 0.6418). The error bars represent standard error of the mean...... 39 2.6 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 50 (30°C: Y=0.0486*X - 1.6685, Z14=3.6, P<0.001, r²=0.6685; 15°C: Y=0.0534*X - 1.4843, Z14=3.8, P<0.001, r²= 0.6544; 4°C:Y=0.0622*X - 0.7251, Z14=3.6, P<0.001, r²=0.6745). The error bars represent standard error of the mean...... 40 2.7 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 63 (30°C: Y=0.0395*X - 1.3973, Z14=3.1, P<0.05, r²=0.5478; 15°C: Y=0.0486*X - 1.3352, Z14=3.6,

P<0.001, r²= 0.4606; 4°C:Y=0.0621*X - 1.1452, Z14=3.9, P<0.001, r²=0.6856). The error bars represent standard error of the mean...... 41 2.8 The average number of ticks alive on day 10 and the corresponding saturation deficit of the jar in which the ticks were kept, the saturation deficit being an integration of the humidity and temperature and defined as the drying power of the air. The error bars represent standard error of the mean...... 42 2.9 The average number of ticks alive on day 20 and the corresponding saturation deficit of the jar in which the ticks were kept, the saturation deficit being an integration of the humidity and temperature and defined as the drying power of the air (Y=-1.098ln(X) + 2.6749, Z44=5.6, P<0.001, r²= 0.4079). The error bars represent standard error of the mean...... 43 2.10 The average number of ticks alive on day 30 and the corresponding saturation deficit of the jar in which the ticks were kept, the saturation deficit being an integration of the humidity and temperature and defined as the drying power of the air (Y=-1.323ln(X) + 1.682, Z44=6.1, P<0.001, r²= 0.5673). The error bars represent standard error of the mean...... 44 2.11 The average number of ticks alive on day 50 and the corresponding saturation deficit of the jar in which the ticks were kept, the saturation deficit being an integration of the humidity and temperature and defined as the drying power of the air (Y=-1.271ln(X) +1.0939, Z44=5.8, P<0.001, r²= 0.6179). The error bars represent standard error of the mean...... 45 2.12 The average number of ticks alive on day 63 and the corresponding saturation deficit of the jar in which the ticks were kept, the saturation deficit being an integration of the humidity and temperature and defined as the drying power of the air deficits (Y=-1.083ln(X) + 0.7739, Z44=5.4, P<0.001, r²= 0.4874). The error bars represent standard error of the mean ...... 46

List of Tables

CHAPTER 2- Effects of temperature and humidity on mortality 2.1 The grams of potassium hydroxide that was added to 100ml of water, and the relative humidity created inside the jar where the ticks were kept after the lid was sealed, detected using an EasyLog USB Data Logger...... 34

CHAPTER 1

Introduction: Ticks and tick-borne disease

1.1 Systematics of ticks

The first scientific description of a tick is thought to have been made by Linnaeus in 1746 (Barker and Murrell, 2004). Today it is considered that there are around 899 species of tick within the subclass Acari and suborder Ixodida, separated into three extant families (Barker and Murrell, 2004): the Ixodidae (hard ticks), the Nuttalliellidae, comprised of a single species Nuttalliella namaqua, and the Argasidae (soft ticks) (Camicas et al., 1998).

Nuttalliella namaqua is thought to be a representative of the last common ancestral tick lineage as it bears characteristics of both the Ixodidae and the Argasidae in addition to some derived features. It feeds on reptiles and is found only in southern Africa, which suggests that this region may have been where ticks originated (Mans et al., 2011).

The Ixodidae have a wide global distribution (Grist, 1992). The family is divided into two groups based on differing morphological traits and, more recently, molecular methods (Charrier et al., 2019). The Metastriata is made up of ~450 species within the genera including Amblyomma, Rhipicephlus (Boophilus), Dermacentor and Hyalomma (Barker and Murrell, 2004). The Prostriata is comprised of ~250 species within a single genus, Ixodes (Charrier et al., 2019). Argasidae, the soft ticks, are composed of ~200 species within 5 currently acknowledged genera; Antricola, Argas, Nothoaspis, Ornithodoros and Otobius. However, the taxonomy of Argasidae is a subject of ongoing debate among taxonomists (Guglielmone et al., 2010). Argasidae are usually found more commonly in areas in which the climate is warm and relatively dry. Consequently, they are not commonly found in abundance in the UK, but instead are widely distributed in the Americas, Africa, and Eurasia.

1.2 The Ixodida: morphology and life cycle

There are multiple anatomical differences between Argasidae and Ixodidae ticks. Soft ticks lack the defining feature of the Ixodidae, the hard sclerotised dorsal plate known as the scutum or conscutum, and instead their dorsal side has a tough, waterproof integument with a mammillated texture. Generally, all nymphal and adult ticks have 4 pairs of legs, specialised mouthparts and bodies comprised of two parts, the posterior idiosoma onto which attach the legs, and the anterior gnathosoma (capitulum) which includes the specialised mouthparts (Needham and Teel, 1991; Sonenshine and Roe, 2014). These mouthparts, made up of the chelicerae, palps, and a hypostome, make ticks distinct among the Acari and are adaptations to their obligate hematophagous behaviour. They are ectoparasites of mammals, reptiles, birds and amphibians and feed on the blood of their hosts by using their paired chelicerae to pierce the dermis and tear the capillaries at the skin surface (Kaufman, 1989; Sonenshine and Roe, 2014).

Fig 1.1 The three-host life cycle of female Ixodidae ticks (Sonenshine and Roe, 2014).

Most ticks go through 4 life cycle stages: egg, larva, nymph, and adult. The larvae have 6 legs and the nymphal stages have 8. Most species have evolved to have a three-host life cycle (Fig 1.1), whereby each instar feeds on a different host before dropping off to moult into the next stage. One and two-host life cycles are less common; moulting from larva to nymph to adult occurs on the same host in one-host cycles, and the adult will locate a new host in two-host cycles. Ixodidae have a single nymphal instar, whereas Argasidae can have several (Oliver, 1989; Sonenshine and Roe, 2014). Ixodidae females have a small scutum, allowing for expansion of the idiosoma as they blood-feed over the course of several days before dropping off the host, laying several thousand eggs in the environment then dying; mating having usually occurred on the host (Sonenshine and Roe, 2014; Estrada-Peña, 2015). In contrast, Argasidae females feed repeatedly and lay between 200-500 eggs between each blood-meal, although this number varies depending upon tick species, size of the female, and the amount of blood ingested. In Argasidae, mating occurs in the environment (Feldman-Muhsam and Borut, 1971; Hoogstraal, 1985; Oliver, 1989). The method used to digest the blood of their hosts also differs. Ixodidae secrete a large proportion of the surplus water from the blood-meal back into the host via the salivary glands. In contrast, Argasidae use the ultrafiltration organ on their coxae, called coxal organs (Estrada-Peña, 2015). Male ticks of most genera usually feed just enough for their reproductive organs to fully develop, although some, such as Ixodes spp., moult from their nymphal form with already active reproductive organs therefore are not required to feed at all. Male ticks remain on the host to mate, which they achieve externally by transferring a spermatheca to the female (Oliver et al., 1974; Sampieri et al., 2016).

Most tick species exhibit endophilic behaviour whereby each life cycle stage shelters in nests, burrows, foliage, or leaf litter until optimum weather conditions permits movement to the top of vegetation to wait for a new host, an activity called questing. The tick latches onto a new host using its front legs and then moves to a suitable location to feed. Questing ticks are attracted to host stimuli, such as body heat, NH3 and CO2. These cues are detected using the Haller’s organ, a highly sensitive pit, packed with chemo-, temperature, and other receptors, positioned on the first pair of legs (Sonenshine and Mather, 1994; Sonenshine and Roe, 2014). Most tick species adopt this sit-and-wait strategy for host location. Only two genera, Amblyomma and Hyalomma, exhibit exophilic behaviour by actively hunting their hosts and moving towards them (Uspensky, 2002; McGarry, 2011). Host-seeking behaviour is dependent upon environmental factors such as changing photoperiod, temperature, humidity, and incident solar energy (Sonenshine and Mather, 10

1994). During questing behaviour, ticks lose water quickly so must rehydrate by descending the vegetation into the litter where humidity levels are higher and water vapour can be absorbed from the atmosphere (Perret et al., 2003).

As soon as humidity conditions are appropriate and/or ticks have hydrated appropriately, questing activity can resume (Estrada-Peña, 2015). When off the host, ticks are in danger of desiccation, predation, starvation, and temperature change (Randolph, 2004a). Questing usually occurs in seasonal patterns, dictated by the climatic conditions required by each tick species and the presence of their specific hosts. Awareness of these patterns is necessary when attempting to assess the risk of pathogen transmission for both humans and animals and for knowing when to apply control measures with greatest effect (Sonenshine and Mather, 1994).

Tick development may cease during its life cycle due to diapause. This is when development ceases seasonally, orchestrated by hormonally controlled mechanisms which allow it to avoid imminent unfavourable conditions. This ensures that oviposition and moulting will occur at optimum summer temperatures. The primary stimulus for inducing diapause within ticks is thought to be photoperiod and temperature (Gray et al., 2016).

1.3 How ticks transmit disease

Ticks are important vectors of a wide range of pathogens globally, due to their high fecundity, environmental resilience, and extensive range of possible hosts, so are of medical and veterinary significance. Among the blood-feeding arthropods, ticks transmit the greatest range of pathogens to host animals (Needham and Teel, 1991); it is thought that an estimated 10% of species are able to transmit pathogens (Jongejan and Uilenberg, 2004). Pathogens are acquired in the blood-meal when the tick feeds from an infected host.

Ticks act as a primary reservoir host and transmission from tick to host occurs primarily when the nymph or adult feeds again. Larvae are usually unlikely to carry these pathogens as they have not previously fed and transovarial transmission from the female to the larvae via the eggs is rare. After acquiring a pathogen in an initial blood-meal, the pathogen will usually be retained throughout all subsequent life cycle stages, with trans-stadial transmission (Bremer et al., 2005).

Tick saliva, secreted from the salivary glands, has many functions including countering host responses through various proteins and inhibitors, maintaining host attachment and control of tick homeostasis, but has a further role in intensifying pathogen transmission between individual ticks on the host via the subcutaneous tissue (Nuttall, 2019). This is called salivary assisted transmission (SAT) whereby transmission is not only to the host via the bite, but to uninfected ticks feeding in a nearby location via saliva (Šimo et al., 2017). This is exacerbated through the gregarious feeding behaviour of ticks on hosts through aggregation-attachment pheromones (Petney and Bull, 1981); minimization of space between feeding ticks facilitates SAT (Randolph et al., 1999; Nuttall, 2019). Additionally, there is evidence that proteins produced within the saliva of feeding ticks increase the likelihood of transmission of some tick-borne viruses (Kubes et al., 1994; Hovius, 2009; Marchal et al., 2010).

1.4 Ticks in the UK

There are over 20 native species of tick found in the UK that feed on mammals, reptiles, or birds (Medlock et al., 2018). The majority of these are hard ticks, the most common being Ixodes ricinus, Ixodes hexagonus and Ixodes canisuga (Abdullah et al., 2016). Other tick species reported include Dermacentor reticulatus, Ixodes frontalis, Haemaphysalis punctata, Hyalomma marginatum and Ixodes arboricola, which are either less common in the UK or introduced through movement of animals across borders (Jameson and Medlock, 2011). Ixodes ricinus, the sheep/deer tick, is the most numerous in the UK and is considered one of the most important threats to public and animal health due to being the primary vector of a variety of pathogens, including Babesia divergens causing babesiosis, Anaplasma spp. causing tick borne fever, Borrelia burgdorferi s.l. causing Lyme borreliosis and Louping ill virus (Medlock et al., 2018).

Ixodes ricinus, also known as the sheep tick, is widespread in deciduous and mixed woodland, moorland, grassland and heathland in the British Isles where the required microclimatic conditions are also met. Ixodes ricinus is a three-host tick, with each instar feeding on a separate host; in the UK immature stages usually feed on smaller mammals and birds and adult ticks parasitise livestock and deer, depending upon the habitat in which the tick is established (Pietzsch et al., 2005). The primary hosts of I. ricinus are livestock, domestic animals such as dogs, humans, small mammals, seabirds, bats and tree-hole and cliff-nesting birds (Jameson and Medlock, 2011). 12

Reproduction occurs on the host, after which females drop off the host into vegetation to lay eggs. Ixodes ricinus lay over 2,000 eggs in a period of up to 6 weeks (MacLeod, 1935). Males and females differ in size and weight, length of the idiosoma and differences in the scutum (Dusbábek, 1996). When unfed, males are between 2.5 to 3mm in length and females are larger between 3 to 4mm, although they can exceed 1cm in length when engorged. Ixodes ricinus is a dark brown/red colour (Walker, 2018a). Ixodes ricinus is known to transmit a multitude of pathogens; those causing babesiosis, Lyme borreliosis, louping ill, tick-borne encephalitis, and anaplasmosis are known to be transmitted by I. ricinus in the UK (Lejal et al., 2019).

Ixodes hexagonus is a three-host tick, widespread across Europe, North Africa, and east to Tajikistan (Walker, 2018b). It is thought that in the UK, populations are mostly established in the south east and are rarely reported in the north (Abdullah et al., 2016). It is difficult to state the exact distribution of I. hexagonus as it is a primarily nest-dwelling species and ticks outside of nests are seldom encountered, thus the actual distribution and abundance of this species is probably greater than present predictions (Arthur, 1951). The primary host of I. hexagonus is hedgehogs, mostly northern white-breasted hedgehogs (Erinaceus roumanicus) and European hedgehogs (Erinaceus europaeus). Ixodes hexagonus is considered to be a red/brown colour and females have a distinct dorsal scutum which is hexagonal shaped. This species is considered somewhat larger than others. Engorged females can exceed 8mm in length and when unfed are between 3.5 to 4mm; females appear lighter in appearance than males. Males are between 3.5 to 3.8mm when unfed (Walker, 2018b). Females spend between 19 and 25 days laying around 1000 to 1500 eggs before dying, and larvae hatch after up to 59 days (Arthur, 1951).

Ixodes canisuga is three-host tick common in central and western Europe, including the UK, and is often misidentified as I. hexagonus (Hornok et al., 2017). It is thought that I. canisuga is more northern in its distribution in the UK, whereas I. hexagonus is more commonly found in the south (Harris and Thompson, 1978). This species is often found on domestic animals, especially dogs, in the UK (Ogden et al., 2000), but has also been observed parasitizing foxes (Harris and Thompson, 1978).

1.5 Dermacentor reticulatus

1.5.1 Morphology and life cycle

Dermacentor reticulatus, first described by Fabricius in 1974, is also known as the ornate cow tick, marsh tick or winter tick (Medlock et al., 2018). It is found on every continent except Australia and is considerably larger than most other ticks found in the UK. A fully engorged female can reach 1cm in length and unfed ticks are between 3.8-4.2 mm in length. Unfed males are slightly larger between 4.2-4.8mm and nymphs and larvae are smaller in size, between 1.4-1.8mm and 0.5mm, respectively (Földvári et al., 2016).

Fig 1.2 Dermacentor reticulatus (A) female (B) male (Medlock et al., 2017).

Morphologically, D. reticulatus most resembles D. marginatus, the only other species of Dermacentor found in Europe, however the latter are generally smaller (Rubel et al., 2016). Dermacentor characteristically have shorter palps than Ixodes, ornate scuta, a palpal spur, and both males and females present white enamel ornamentation, with male ornamentation covering the entire scutum (Fig 1.2). Identifying immature instars is difficult as they resemble juveniles of other genera (Földvári et al., 2016). There are 36 recognised species within the genus Dermacentor (Wang et al., 2019).

Compared to more common UK ticks, such as I. ricinus, the ecology of D. reticulatus is less well known; there is a particular paucity of data relating to the immature instars which are almost never collected in blanket-drags. Dermacentor reticulatus has a three-host life cycle which takes approximately 1-2 years to complete (Rubel et al., 2016) and it has been suggested that its 14

development from larva to nymph is more rapid than other tick species such as I. ricinus (Randoph et al., 1999). Dermacentor females usually oviposit between 3,000 and 7,500 eggs, depending on the species, and the degree of engorgement. As with other tick species, the eggs are covered with secretions from the Gené’s organ to prevent desiccation (Sonenshine and Mather, 1994; Šimo et al., 2004). High levels of mortality in the environment, thought to surpass 99.9% within the larval and nymphal instars, limits increases in abundance and distribution (Sonenshine and Mather, 1994). In contrast to I. ricinus, the larvae and nymphs are thought to exhibit nidicolous behaviour, remaining in the burrows or nests of their hosts (Pfäffle et al., 2015). Although not currently proven, it is thought that larvae may remain deep in leaf litter or within burrows. This theory is supported by the absence of larvae from blanket drags. When they are recovered, the immature stages of D. reticulatus are found on small mammal hosts, often in mid-summer, with oviposition presumed to have occurred in spring (Pfäffle et al., 2015). Egg hatch occurs 12-19 days after oviposition (Nosek, 1972). Larvae feed for 2.5-6 days and nymphs generally for a longer period, around 4-12 days. The activity of the adults generally begins in late August/September through to April/May, only disrupted by changes in climatic variables such as falls in temperature (Rubel et al., 2016). Throughout the winter, D. reticulatus undergo hormonally controlled diapause to halt development during the unfavourable colder conditions (Gray et al., 2016). It has been suggested that D. reticulatus begins activity after diapause earlier in spring than other species because they are able to withstand cooler temperatures (Tokhov et al., 2014). Unusually for a three-host tick species, overwintering has been observed for D. reticulatus on a host (Karbowiak et al., 2003).

Dermacentor reticulatus quests and will attach to a wide range of hosts (Sonenshine and Mather, 1994). It is thought that different species act as the primary host depending on the type of habitat and geographical location of the tick population. The most common hosts in the UK include dogs, horses, cattle and, rarely, humans (Medlock et al., 2017). The average questing height is 55cm (Menn, 2006) and possible hosts are detected using odour cues. For this reason, D. reticulatus are often found alongside paths used by domestic dogs and wildlife (Tharme, 1993). Male ticks remain on a host for a longer period than females, around 2-3 months, in order to mate with females, thus female ticks are more frequently observed questing than males (Tharme, 1993; Kiszewski et al., 2001). When feeding, adult D. reticulatus ticks aggregate causing visible wounds with local inflammation (Buczek et al., 2015). This aggregation is deemed to be a consequence of

the production of aggregation-attachment pheromones, as demonstrated in other ticks (Petney and Bull, 1981) but not yet investigated in D. reticulatus.

1.5.2 Distribution

Dermacentor reticulatus is thought to be among the top three most abundant tick species in central Europe, along with D. marginatus and I. ricinus (Rubel et al., 2014; Rubel et al., 2016). However, it is relatively rare in the UK; a recent national survey of ticks on dogs in the UK detected only three D. reticulatus infestations from a sample of over 8,000 (Abdullah et al., 2016). It is thought that the distribution of D. reticulatus in Europe (Fig 1.3) is divided into two genetically distinct groups, with a Western European population and an Eastern European population, with an overlap of the two groups in Poland (Paulauskas et al., 2018). This tick is found in much of continental Europe including, but not exclusively in, Hungary (Földvári and Farkas, 2005; Sréter et al., 2005), Czech Republic (Nosek, 1972; Hubálek et al., 2003), Germany (Menn, 2006; Rubel et al., 2014; Pfäffle et al., 2015), Slovakia (Nosek, 1972; Bullová et al., 2009; Špitalská et al., 2012), Poland (Bartosik et al., 2011; Wójcik-Fatla et al., 2011; Biernat et al., 2014; Mierzejewska et al., 2015b; Mierzejewska et al., 2016), Switzerland (Perret et al., 2000), Italy (Olivieri et al., 2017), Russia (Tokhov et al., 2014) and France (Martinod and Gilot, 1991).

It has been suggested that the distribution of D. reticulatus is expanding north through Europe (Rubel et al., 2016). Newly established populations have been observed in Poland (Zygner et al., 2009), Germany (Dautel et al., 2006) and The Netherlands (Matjila et al., 2005).

Fig 1.3 Current D. reticulatus geographical distribution in Europe detailing where D. reticulatus is present, has been introduced, and is absent based on data provided by the Vector-Net project and published historical data (from ECDC and EFSA, 2019)

In Europe, it has been suggested that D. reticulatus is largely associated with alluvial forest and swamp habitats, and the species is able to survive submergence for extended periods of time (Rubel et al., 2016). Eggs have even been observed to survive in basins of rainwater (Hoogstraal, 1967). However, established populations may also survive in dry conditions, such as on heathland and grassland, and are rarely observed in dark or coniferous forests, unlike I. ricinus which is more tolerant of closed, forested environments (Cerny et al., 1982; Rubel et al., 2016).

Fig 1.4 Current known distribution of Dermacentor reticulatus in the United Kingdom using data from Public Health England, the Big Tick Project and field collections between 2009-2016. The symbols show where Dermacentor reticulatus populations were reported to be by the mentioned sources (Medlock et al., 2017).

The north-western border of the current distribution of D. reticulatus is the UK. It is thought that D. reticulatus has been present in small populations in the UK for over 100 years according to the Health Protection Agency (HPA) and British Records Centre (BRC) recording scheme (Medlock et al., 2011), although the species may have been present for longer than this. According to historical records, questing D. reticulatus have been recorded in every month of the year in the UK, primarily during September to June with an activity peak in March and April (Medlock et al., 2017).

In the UK, D. reticulatus is thought to be associated primarily with maritime grassland and sand dune habitats, including grazing marsh, and has been recorded in coastal sites in Devon and Wales (Medlock et al., 2017). Historical and current records also suggest that many of the sites in the UK where this species is found are old rabbit warrens (Medlock et al., 2017). Recently, D. 18

reticulatus has become more common in Wales and southern England (Fig 1.4). There are now four confirmed UK populations; the coast of West Wales between Harlech and Borth where their habitat is dominated by marram grass (Ammophilia arenaria) and Burnet Rose (Rosa pimpinellifolia), within parks and coastal grassland in Essex around Potton Island and Harlow, and on both the North and South Devon coasts namely in the Bideford sand dunes and Bolt Tail grassland (Medlock et al., 2017; Medlock et al., 2018). A population near Harlow in Essex, found recently in an area of grassland, is thought to be relatively newly established; it has been suggested that this may be linked to the movement of livestock and dogs. It has been suggested that D. reticulatus may continue to spread in this area (Medlock et al., 2017). The establishment of this population suggests that the presumed association between D. reticulatus and coastal dune grassland in the UK may be incorrect. Alternatively it may be that the Harlow population is only transitory and will die out.

1.5.3 Epidemiological Impact

Dermacentor reticulatus is of increasing interest due to its rising epidemiological significance worldwide. It is suggested that there are over 40 different pathogens, known to infect both animals and humans, that have been detected in D. reticulatus, globally (Földvári et al., 2016). This tick species is thought to be an effective vector due to its high reproduction rate, rapid development rate, promiscuous feeding, and ability to survive in adverse conditions (Földvári et al., 2016).

Pathogens transmitted by D. reticulatus in Eurasia include tick-borne encephalitis virus, Rickettsia spp., Anaplasma spp. (Zivkovic et al., 2007), Babesia spp., Theileria equi, Kemerovo virus, bluetongue virus, Borrelia spp., and Francisella spp. (Földvári et al., 2016; Rubel et al., 2016). When considering the future threats from of veterinary disease that could become problematic in the UK, the most significant pathogens carried by D. reticulatus are Babesia canis, Theilaria equi and Babesia caballi. Additionally, Rickettsia raoultii has recently been detected in D. reticulatus populations in the UK (Tijsse-Klasen et al., 2011, 2013).

1.5.4 Risks to veterinary and public health

Canine babesiosis

The apicomplexan piroplasmid parasite Babesia canis is the most notable pathogen spread by D. reticulatus and the causal agent of canine babesiosis, also known as piroplasmosis. Infection in dogs’ results in a range of clinical signs, including fever, jaundice, anaemia, inflammation and, in severe cases, organ failure and death (Irwin, 2009). Babesia canis is capable of transovarial and transstadial transmission within ticks (Gray et al., 2019), allowing D. reticulatus populations to act as effective reservoir hosts, maintaining this pathogen in the environment. The evidence suggests that wildlife cannot act as reservoirs, however domestic dogs can. Thus, it is thought that dogs are capable of transmitting the pathogen to feeding ticks and are, at least partially, responsible for its continuing dispersion. Vertical transmission of B. canis from female dogs to their litter has recently been observed, which intensifies the spread (Mierzejewska et al., 2014). Babesia canis was detected for the first time in the UK in 2016, in a population discovered in Harlow, Essex (Swainsbury et al., 2016). Subsequently, the prevalence of infected D. reticulatus ticks at this site was shown to be over 85% (de Marco et al., 2017). It is likely that an infected dog introduced B. canis in this area and that the population of ticks at Harlow have sustained the pathogen via transstadial and transovarial transmission (Mierzejewska et al., 2014).

Babesia canis was only isolated from a single D. reticulatus tick during a study in which 4750 ticks sent in by veterinary practices across the UK were examined for pathogens (Abdullah et al., 2018). However, this tick came from a dog which had recently travelled back from France, emphasising the threat posed by untreated travelled dogs importing pathogens from other countries, including B. canis. This presents a serious veterinary health problem in the UK (Medlock et al., 2018).

Equine piroplasmosis

Dermacentor reticulatus is a known vector of equine piroplasmosis caused by Theileria equi and Babesia caballi. These cause disease in equids, especially horses. Both pathogens can cause anaemia, neonatal death, and a high risk of mortality among other clinical signs (Guidi et al., 2015). Recovered horses can act as reservoirs, allowing the further transmission to ticks. Sexual

development of T. equi and B. caballi occurs within the tick mid-gut and in its salivary glands. Both transstadial and transovarial transmission occurs, resulting in persistence within tick populations. Therefore, the distribution and spread of these pathogens is closely associated with the distribution and spread of their tick vectors (Scoles and Ueti, 2015). Despite protocols, the movement of infected horses across borders contributes considerably to the continuing spread of these pathogens, exposing them to naïve tick populations in new regions (Guidi et al., 2015). Historically, equine piroplasmosis has not been detected in the UK and only small populations of suitable vectors have been present. Hence, B. caballi and T. equi have been of little concern. However, the recent increased number of recordings of D. reticulatus in the UK suggests that the potential for the status of equine piroplasmosis to change is high, should the pathogens be inadvertently introduced. A recent study concluded that horses imported to the UK are not monitored carefully enough; given that 8% of submitted equine samples tested positive for pathogen DNA, the proportion of infected animals was likely to have been considerably higher, particularly given the lack of sensitivity of the PCR test available (Coultous et al., 2019).

Rickettsia spp.

Rickettsiae are intracellular bacteria with more than 20 confirmed species, 14 of which infect humans (Tijsse-Klasen et al., 2011), causing a variety of syndromes such as tick-borne lymphadenopathy (TIBOLA) (Silva-Pinto et al., 2014). The genus is separated into 3 distinct groups, one being the spotted fever group (SFG), the majority of which are transmitted by ixodid ticks, transmission occurring during feeding, but also transovarially and transstadially. Tick vectors of Rickettsia spp. have been identified on migratory birds in the UK, indicating that the long-distance transportation of ticks on migratory birds has the potential to introduce new pathogens into UK tick populations (Graham et al., 2010). One study found that 27% of ticks tested were infected with Rickettsia spp., the majority being Rickettsia raoultii, in D. reticulatus collected from a variety of locations across the UK (Tijsse-Klasen et al., 2011). Currently, SFG rickettsiae in D. reticulatus have been detected in Essex and Wales, but they are likely also to be present elsewhere in the UK (Tijsse- Klasen et al., 2013). Rickettsia slovaca has been confirmed as present in the UK (Pietzsch et al., 2015) and D. reticulatus has been identified as a possible vector (Špitalská et al., 2012; Földvári et al., 2013). The presence and establishment of these pathogens in the UK and the ability of D.

reticulatus to act as a vector and reservoir host suggests that Rickettsia spp. may pose serious public and veterinary health risks.

Tick-borne encephalitis virus

Tick-borne encephalitis virus (TBEV), an infection of the central nervous system, poses a public health threat in Europe and the majority of Asia. TBEV is classified within the genus of Flavivirus; it causes chronic fever, muscle pains, fits, exhaustion, paralytic disorders in its host and in rare cases infection can be fatal, depending on the form of TBEV and the age and genetics of the host (Gritsun et al., 2003).

TBEV persists in ticks through transstadial, transovarial, viraemic and salivary assisted transmission (Gritsun et al., 2003). Currently, the role of D. reticulatus in TBEV transmission is considered to be less important than that of I. ricinus. However, D. reticulatus has been implicated as a potential reservoir host of TBEV in Poland. Biernat et al. (2014) found a TBEV infection rate within D. reticulatus varying between 1 and 12% in all locations studied. An earlier study had found a 10.8% infection rate in D. reticulatus, which was higher than observed in I. ricinus (Wójcik-Fatla et al., 2011). It is thought that D. reticulatus is more prevalent on cattle than I. ricinus and that cattle act a reservoir of TBEV, thus exacerbating the spread of TBEV by D. reticulatus (Mierzejewska et al., 2015b). Recently TBEV has been detected in ticks in the UK for the first time in an area bordering Dorset and Hampshire (Public Health England, 2019). Though the current risk to public health is deemed low, the establishment of TBEV among cattle and D. reticulatus populations could make TBEV in the UK a considerable public health threat.

1.5.5 Environmental requirements

Nearly all species of tick are adapted to a specific climatic window, associated with particular habitats (Sonenshine and Mather, 1994). Optimal environmental conditions considerably improve the survival and development rate of ticks off-host (Needham and Teel, 1991). Humidity and temperature are particularly important, as these two variables together determine the rate of desiccation (Estrada-Peña, 2015). Together temperature and humidity can be expressed as an integrated measure, saturation deficit, representing the drying power of the atmosphere.

Saturation deficit is the amount by which the water vapor in the air must be increased to achieve saturation at any given temperature (Randolph and Storey, 1999). The amount of water air will hold at saturation increases with rising temperature and thus the same relative humidity will correspond to a greater saturation deficit at warm temperatures. Saturation deficit provides important information about the amount of water available in the atmosphere to ticks at any given temperature.

Dermacentor reticulatus has been observed to have a high tolerance to starvation; it has been suggested that adults can survive longer than 3 years without a host (Földvári et al., 2016) and for 2.5 years post-moulting (Razumova, 1998). Thus, it is thought that tick abundance is controlled primarily by abiotic factors, secondarily by predation.

Hosts

Host availability determines the presence or absence and abundance of tick populations, as suitable hosts are required for successful tick development, feeding and reproduction (Estrada- Peña, 2015). A range of different species have been recorded as hosts for D. reticulatus, depending on climate and habitat type. There are four main locations in which D. reticulatus has been recorded in the UK, particularly along the coast of Wales (Medlock et al., 2017). Past studies have found D. reticulatus populations close to camping grounds at Shell Island and in the vegetation alongside paths near golf courses where dogs often visit at Morfa Harlech, suggesting that dogs might be important hosts for questing adult ticks (Evans, 1951). This supports observations by Tharme (1993) that D. reticulatus is often found exhibiting host-seeking behaviour on the margins of habitat where there are high densities of suitable hosts, such as popular dog walking paths. It is thought that areas such as these have a higher density of ticks due to the tendency of host animals to repeatedly use particular paths or sites (Estrada-Peña, 2015). It is notable that grazing livestock are also present in all four known D. reticulatus locations in the UK, indicating that large mammals such as cattle and horses may also be hosts for adults (Medlock et al., 2017).

Hosts for D. reticulatus larvae and nymphs in the UK are thought to include voles, moles, shrews, hares, hedgehogs, birds and rabbits while nymphs have also been recorded on weasels, polecats, cervids, goats and dogs (Nosek, 1972; Földvári et al., 2016).

Photoperiod

Photoperiod is considered an important factor for inducing various behaviours in D. reticulatus (Tharme, 1993; Bartosik et al., 2011). Mild winters may be insufficient to initiate winter diapause in Europe, which is thought to be triggered by declining temperature (Tharme, 1993). Additionally, photoperiod may stimulate the beginning of a summer diapause and then reactivate questing activity in August and September, as questing begins before the temperature decline, thus an increase in daylight hours is thought to be the primary stimulus (Tharme, 1993). Photoperiod both prevents and stimulates host-seeking activity of unfed D. reticulatus, as well as initiating a cessation of development in engorged ticks, enabling survival throughout adverse weather conditions (Bartosik et al., 2011; Estrada-Peña et al., 2013). There is also evidence that some ticks, such as I. ricinus, are active in the dark; one study found that only 1% of all movement occurred during maximum light intensity (Perret et al., 2013). Therefore, it is possible that light intensity may have some similar influence over D. reticulatus activity, although this has not yet been investigated. Movement, primarily during darkness, could be a behaviour that limits tick desiccation (Perret et al., 2013) while allowing questing activity to coincide with that of their hosts.

Humidity

Maintenance of body water is vital for terrestrial arthropods due to a large surface-to- volume ratio, making them particularly prone to desiccation. The relative abundance of host- seeking ticks is impacted by a multitude of factors, the most influential of which is thought to be relative humidity (and saturation deficit); changes in the numbers of questing ticks are strongly correlated with changes in saturation deficit (Randolph and Storey, 1999; Perret et al., 2000; Randolph et al., 2002; Hubálek et al., 2003; Perret et al., 2003). Ticks have a multitude of mechanisms guarding against water loss in desiccating environments, including the integument which acts as a highly efficient physical barrier against water loss (Meyer-König et al., 2001b). Their respiratory system shielded from the environment via spiracular valves and the excretion of nitrogenous waste in the form of guanine crystals also limit water loss. Rehydration is achieved through vapour absorption, feeding and via metabolic activity (Needham and Teel, 1991). Tick saliva also has a function in water balance, containing a hygroscopic substance that absorbs water 24

vapour from the atmosphere (Nuttall, 2019). During host-seeking activities, water loss and energy consumption are increased thus limiting survival. It is imperative that ticks can survive until a suitable host is found (Meyer-König et al., 2001a). Should the relative humidity fall so that ticks are unable to absorb water efficiently from the atmosphere, the resultant desiccation increases the mortality rate. In more humid conditions therefore, ticks are able to quest for longer as the desiccation rate is slower (Estrada-Peña et al., 2013). The highest water loss tolerance recorded for any tick species is for Amblyomma americanum, the lone star tick, which can lose as much as 70% water mass before loss of locomotor ability occurs (Edney, 1977). Most ticks are only able to move vertically, so when they experience high levels of water loss while questing they move back down to the vegetation mat to rehydrate; frequent movement between vegetation interspersed by bouts of questing have been recorded in low humidity conditions. This movement requires energy from the tick’s fat reserves which is restored by blood-feeding. Immature stages usually quest at lower heights within vegetation because they are more susceptible to desiccation and this may be associated with their larger surface area to volume ratio, higher metabolic rate, and smaller reserves of fat (Mejlon and Jaenson, 1997; Randolph and Storey, 1999).

It has been suggested that D. reticulatus may be more sensitive to desiccation than the more common I. ricinus and that it therefore has a strong association with bodies of water in the environment (Nosek, 1972; Hubalek et al., 2003). In contrast, some research has suggested that D. reticulatus may be more resistant to desiccation than Ixodes spp., surviving for significantly longer periods in dry air before dying (Lees, 1946). It is thought that D. reticulatus require relative humidities exceeding 80% to permit active uptake of water vapour from the atmosphere (MacLeod, 1935; Meyer-König et al., 2001b). There is limited research on whether age or sex have any influence on desiccation ability, however there is some evidence suggesting that the permeability of the tick integument increases with age (Needham and Teel, 1991) and that females have a higher rate of survival in drier environments than males (Meyer-König et al., 2001b). For I. ricinus, it is thought that the relative humidity must exceed 70-80% for survival (MacLeod, 1935). Other literature estimates that ticks will stop losing water via evaporation when the atmospheric moisture level reaches 92% RH (Lees, 1946).

In studies in eastern Europe, D. reticulatus mortality was found to increase with higher saturation deficit (Meyer-König et al., 2001a) and activity was highest when the relative humidity

was at its highest, between 55-65% (Buczek et al., 2017). In contrast, humidity was reported not to have a significant impact on the activity of D. reticulatus in eastern Poland, where temperature and photoperiod were the only variables correlated with activity levels (Bartosik et al., 2011). Hence, clearer information is required on the relationship between relative humidity and survival of D. reticulatus. Most previous work on D. reticulatus has been undertaken in Eastern Europe, so it is important to also investigate this behaviour in the UK.

There is some debate on the impact of rainfall on tick activity and desiccation. Several studies based in Europe found no correlation between rainfall and D. reticulatus activity (Martinod and Gilot, 1991; Hubálek et al., 2003), although both studies used weather data supplied by a local meteorological station instead of recording temperature of the vegetation where the ticks were found. In contrast, it has been suggested that a higher average rainfall inevitably results in a higher humidity, thus resulting in better conditions for D. reticulatus survival (Gerstengarbe et al., 2008, Süss et al., 2008). It has also been argued that rainfall increases the activity of D. reticulatus after diapause (Olivieri et al., 2017).

Temperature

Temperature has been shown to have a significant effect on D. reticulatus activity and mortality in France and Poland (Martinod and Gilot, 1991; Bartosik et al., 2011). However, a weakness of many studies is, as mentioned above, that they use data from local weather stations, which may not be a true representation of the temperature of the vegetation layer close to the ground where the ticks are found (Estrada-Peña et al., 2013; Boehnke et al., 2017). Vegetation temperature has been found to have a significant impact on the activity of D. reticulatus as it influences the saturation deficit, thus affecting activity and mortality rates (Tharme, 1993; Estrada- Peña, 2015).

In western Siberia, the most easterly region of the known range of D. reticulatus, adults have been observed host-seeking for only 3 months during spring and 3 months during the summer. Diapause, or quiescence, halts development during unfavourable conditions (Gray et al., 2016). Thus, questing ticks are not found in these areas between September and April, due to the harsh winter temperatures. In the UK, the most westerly distribution of this species, D. reticulatus appears to have a short summer diapause between June and August, and this is thought to occur 26

due to unfavourably warm temperatures and low humidity. The brief winter period when ticks are not found questing is probably not a true diapause, as adult D. reticulatus can reappear very quickly when the weather becomes milder and they can sometimes continue questing throughout winter (Tharme, 1993). Occasionally during severe winters, long periods of heavy snow cover on top of the litter provides protection, insulating the ticks against low temperatures (Estrada-Peña, 2015).

Dermacentor reticulatus has been shown to be active during the winter months in temperatures that are too cold for I. ricinus, suggesting that D. reticulatus is cold-hardy compared to other ixodid tick species (Tharme, 1993). The threshold temperature for I. ricinus activity is thought to be 7°C (Tharme, 1993; Perret et al., 2000). Active questing of D. reticulatus ticks has been observed at temperatures as low as 1°C and a soil temperature of -0.1o C in mainland and Eastern Europe (Nosek, 1972; Hubálek et al., 2003). While monitoring questing at a site in Wales over a period of 24 hours, the lowest temperature recorded at which D. reticulatus was still questing at was 3.3°C and the overnight temperature was -5.4°C, and the surface of the sand at the marsh site was frozen (Tharme, 1993). However, this latter study only monitored this activity over a single night and previous studies have primarily used data from weather stations rather than the vegetation layer. While using weather station data is accurate to some extent, as atmospheric conditions affect the conditions in the vegetation and affect ticks during questing, it is thought that using data recorded at vegetation height is preferred, as this is where ticks rehydrate. If the climate with the vegetation is unsuitable, then the ticks cannot survive in that location. In this way, the temperature threshold for D. reticulatus questing activity remains unclear. The majority of studies on D. reticulatus are based in eastern and central Europe, mostly in Poland, Czech Republic and France, leaving the activity and mortality of this species under different temperature conditions undetermined in the UK.

1.5.6 Changing abundance and distribution

Over the past decade, many arthropod vectors, including D. reticulatus, are thought to have increased their abundance and expanded their distribution, despite on-going research and expenditure on ectoparasiticides. These changes are thought to be the result of the synchronous impact of many factors, creating more favourable environments for tick establishment (Karbowiak, 2014; Paulauskas et al., 2018). 27

There are a multitude of elements that may influence the dispersion of D. reticulatus such as human migration, globalisation, land-use change, trade, and urbanisation (Karbowiak, 2014; Medlock et al., 2015). Change in agricultural practices, such as a reduction in the use of pesticides, abandonment of agricultural land, and more reforestation can be favourable for D. reticulatus and other tick species (Karbowiak, 2014, Mierzejewska et al., 2015a). There has also been a reported increase in tick abundance in urban greenspace especially, such as parks and small woodlands (Medlock et al., 2018). These, along with other similar spaces contain a mosaic of vegetation and are an increasingly suitable habitat for the hosts of D. reticulatus. Thus, ticks that are dispersed by their hosts to new habitats are more easily able to establish a population (Karbowiak, 2014). Movement of animal hosts between sites using wildlife corridors such as river valleys, more regular travel to the countryside with domestic dogs, and an increase in livestock trading routes exacerbates the issue (Mierzejewska et al., 2016; Rubel et al., 2016; Medlock et al., 2018).

Climate change is often highlighted as a key driver of changes in tick distribution globally and includes a multitude of variables such as changing patterns of temperature, wind direction and precipitation. Throughout Europe mean temperatures have increased, more rainfall has been recorded and seasons of optimum growth have lengthened resulting in a warmer, wetter northern Europe (Karbowiak, 2014). Thus, conditions are more favourable for tick species that were previously incapable of surviving in a cooler, more temperate environment. For instance, Hyalomma marginatum, the primary vector of Crimean-Congo haemorrhagic fever virus in Europe, enters the UK on migratory birds every year, however the climate has previously been too cold for it to survive. It has been suggested that in the near future, the UK could become warm enough for this tick to become established (Gale et al., 2012).

1.6 Study aims

The aim of the research described in this thesis was to determine the effect of temperature and humidity on the survival of the tick D. reticulatus collected from UK populations. Ticks were to be collected from various sites on the west coast of Wales, north and south coasts of Devon, and Essex. As highlighted in the review above, the environmental constraints that determine the distribution and abundance of this species are poorly understood and very few studies have investigated D. reticulatus in the UK. Since this tick is an important vector of the causal agents of a 28

variety of diseases of considerable veterinary and public health concern, identifying its basic microclimatic requirements is vital. The data provided by this research will aid in analysis of the risk from D. reticulatus in the UK and provide information on the potential spread of emerging tick- borne pathogens. This knowledge will be useful for future population modelling vital for sustainable management of this vector.

CHAPTER 2

Effects of temperature and humidity on mortality

2.1 Introduction

As discussed in Chapter 1, the abundance and distribution of D. reticulatus is increasing, both in central Europe and in the UK (Matjila et al., 2005; Dautel et al., 2006; Zygner et al., 2009; Rubel et al., 2016; Medlock et al., 2017). Climate change is thought to be a key driver in this changing pattern (Karbowiak, 2014). In the future, D. reticulatus may become more widely dispersed and abundant in the UK (Gray et al., 2009; Jameson and Medlock, 2011) and the threat the pathogens it transmits may become more substantial (Földvári et al., 2016; Rubel et al., 2016). Humidity and temperature are thought to be the most influential environmental factors affecting survival and abundance because together they determine the rate of desiccation (Randolph and Storey, 1999; Bartosik et al., 2011). However, there is a lack of data or consensus on the effects of these factors on D. reticulatus. Therefore, the aims of this study were to examine the mortality rate of D. reticulatus collected from four locations under varying temperatures and humidities to better understand the basic climatic requirements and the effect that variations in these might have on the mortality of D. reticulatus in the UK as a whole.

2.2 Methods

2.2.1 Collection sites

There are four known areas where D. reticulatus can be found in the UK; West Wales, Essex, North Devon, and South Devon (Medlock et al., 2017). Ticks for this study were collected from sites within all four areas between September 2019 and February 2020.

Tick collection in Wales occurred along a stretch of coastline near Harlech. Harlech is located on the west coast of Wales within Snowdonia National Park. Tick collection took place in three areas. Behind Harlech beach there is a dune area at 52.8731° N, 4.1294° W containing some small bodies of water, grazing cattle and dog-walking paths. The dunes nearest the beach are 30

mostly devoid of plant life other than marram grass (Ammophila arenaria). The majority of ticks were found in the dune slacks further from the beach where there are a variety of plant species and small bodies of water on flatter ground. The second collection site was around 5km south of Harlech on Shell Island at 52.8080° N, 4.1443° W, a peninsula on which there is a 1.2² km campsite during the summer months and during the winter months local farmers graze sheep. Access to the island is only possible along a causeway across the River Arto estuary. Shell Island attracts many dog-walkers in the summer months. The majority of ticks were collected south of the main campsite in vegetative areas surrounding large sand dunes. The third collection site was in the dune area near Dyffryn Ardudwy at 52.7813° N, 4.1182° W, located in the dunes next to Benar Beach. Tick collection occurred primarily in sheltered, highly vegetative areas positioned behind the large sand dunes. Dog-walking trails are numerous here and cattle are grazed in pastures just behind the dune area in which the ticks were collected.

Braunton Burrows and Northam Burrows are on the north coast of Devon, west of Barnstaple. Ticks were collected from Braunton Burrows, a dune area of approximately 1400 hectares at 51.0928° N, 4.207895° W behind Saunton Sands beach. Across the river mouth to the south is Northam Burrows Country Park at 51.0545° N, 4.2245° W. Ticks were collected across both locations, in scrubland beyond the beach. Both areas are popular with dog-walkers and farm animals are periodically grazed there, usually cows in Braunton Burrows and horses in Northam Burrows.

Hope Cove is located on the south coast of the UK in Devon. Tick collection took place on the cliffs next to Hope Cove near Bolt Tail at 50.2427° N, 3.8660° W. There are numerous walking trails popular with dog-walkers. The ground on top of the cliffs is primarily covered by short grass; thicker, taller vegetation is located near the edges of cliffs looking over Hope Cove beach. This is where the majority of ticks were found.

Old Hall Marshes, an RSPB nature reserve at 51.7763° N, 0.8542° W, was the site in Essex where ticks were collected. This area contains many bodies of water, extensive saltmarsh, reedbeds and vast grazing marshes. The reserve is open to the public and is popular with dog-walkers.

The locations ticks were collected from differ in climate. The mean daily minimum temperature in January varies between sites. On Welsh coasts this temperature ranges between 31

3°-4°, whereas in eastern England it is 1°, and in Devon it’s thought to vary between 1° -5° (Met Office UK, 2020). In contrast, in July, the mean daily maximum temperature is 18° on the Welsh coast, between 20-23° in Essex and upwards of 19° in the South-West. Along with temperature, rainfall also varies between the locations, from 1000mm annually along the Welsh coast, 900- 1000mm on the south-west coast to less than 700mm annually in the east of England, sometimes as little as 500mm in the driest areas (Met Office UK, 2020).

2.2.2 Collection method

All ticks were collected using the standard blanket-dragging technique. A bamboo pole of approximately 1.2m had a 1m² white cotton blanket attached. On each end of the bamboo pole green twine was tied and used to pull the apparatus along the ground and over vegetation. The twine was approximately 2.5m long. The blanket acted as a makeshift ‘host’, the movement and texture imitating that of a roaming mammal. The blanket was dragged along the ground slowly and turned over to check for tick attachment. The usual protocol while collecting ticks such as Ixodes ricinus is to check for ticks every 10m, however D. reticulatus is a larger species and is easily knocked from the blanket. To limit this occurrence, the blanket was turned over every 5m. D. reticulatus were identified using key morphological features, primarily the white enamel ornamentation and the size of the tick. Forceps were used to remove the tick from the blanket and place it into collection tubes (Sarstedt, 30ml). Into each collection tube a blade of grass was placed for ticks to attach to during transport. There were normally between 15 and 30 ticks stored in each tube (Fig 2.1). If the collecting trip spanned 2 days, those collected on the first day were kept in a portable 4°C mini-fridge over-night. Once returned to the laboratory, the ticks were kept in their original collection tubes in a 4°C fridge until the experiment was set up. Each tick was used only once, and ticks remained in the fridge at 4°C for no longer than 3 days.

All D. reticulatus ticks collected were adults. As discussed in Chapter 1, larvae and nymphs are difficult to find in their natural habitats, as it is believed that immature stages remain in host burrows or on small mammal hosts. Other tick species collected on the blankets were not used in the study.

Fig 2.1 Collection of ticks: a) a male D. reticulatus on the blanket, having attached while the blanket was being dragged over vegetation; b) the blanket used for tick collection, made up of a 1.2m bamboo pole and 1m2 cotton sheet; c) D. reticulatus, a male and a female, within a collection tube ready for transportation to the laboratory

2.2.3 Experiment design

During the study, glass desiccator jars (2.4L, VWR International) were used. Within each jar 3 tubes (50ml, Sarstedt) were placed and within each tube were 5 ticks of mixed sex and several ‘grass stems’ made out of filter paper (Whatman No. 1) for tick attachment. The lids of the tubes were removed and the top covered with fine mesh, secured with a rubber band allowing airflow between the jar and the ticks in the tube (Fig 2.2). Five different relative humidities (RH), 20% RH, 40% RH, 60% RH, 80% RH and 95% RH, were created by altering the concentration of potassium hydroxide KOH added to 100ml of water (Table 2.1) (Solomon, 1951). The solution was poured into the bottom of the jar and the tubes were placed on a mesh covering, preventing the tubes of ticks

falling into the KOH. Jars were placed in incubators set at 4°C (Leibherr, Medline), 15°C and 30°C (MLR-351H; Sanyo, Panasonic, Loughborough, UK).

The lid rim was covered by a fine layer of Vaseline to create a seal and once the KOH had been poured in and the ticks placed inside, the lid was quickly placed on the jar (Fig 2.2). The jar was then put in the incubator and remained there for 9 weeks (63 days) or until all the ticks inside were dead. Tick mortality was checked every 2 or 3 days and recorded. When the ticks were checked for mortality, the lid was replaced as quickly as possible after tubes were removed or put back in to ensure as little change in the internal humidity of the jar as possible. An EasyLog USB Data Logger (Lascar Electronics) was used to check that the humidity was correct in each jar and the KOH solution was changed every 2 weeks to ensure the humidity stayed at the desired level. The incubators used were kept dark inside during the investigation for consistency and so that light did not impact tick activity levels. The experiments were completed in stages between 4th December and 8th April (2019/2020), each one starting immediately after returning with new ticks from a field trip and continuing for the 9-week period. Overall, 225 D. reticulatus ticks were used in the study.

Table 2.1 The grams of potassium hydroxide that was added to 100ml of water, and the relative humidity created inside the jar where the ticks were kept after the lid was sealed, detected using an EasyLog USB Data Logger.

KOH (grams) Humidity RH%

100 20

80 40

40 60

15 80

6 95

Fig 2.2 Desiccator jar assemblage: a) 5 ticks and a filter paper ‘grass stem’ within a tube; b) fine mesh to allow air flow was secured on top of a tube with a rubber-band to hold it in place; c) two tubes within a desiccator jar standing on top of mesh to prevent them falling into the solution below; d) a desiccator jar containing tubes of ticks with the lid sealed on top using Vaseline to keep the internal humidity from changing

2.2.4 Statistical analysis

During statistical analysis, the primary dependent variable was the number of dead ticks recorded. Poisson regression and a generalized linear model (GLM) were used to investigate how tick survival varied with humidity, temperature, and time. In the GLM temperature and humidity were inputted as continuous variables and count was the dependent variable; time was entered as a random factor to control for repeated measures. Data were analysed using RStudio (R, version 3.4.2, R Core Team 2017). Temperature and humidity were used to calculate the saturation deficit (Randolph and Storey, 1999) and the relationship between saturation deficit and count was investigated by Poisson regression. The Poisson distribution was found to be the best fit for dead tick count, and the model fulfilled the suitability requirements (residual deviance less than twice the residual degrees of freedom, no over or under-dispersion and normal distribution of standardised residuals). The ticks were exposed to the differing climatic variables for 63 days, so 5 time points were analysed using R Studio: day 10, day 20, day 30, day 50 and day 63.

2.3 Results

There was a significant interaction between humidity, temperature, and time on tick survival overall (Z214=3.8, P<0.001) so each time point was then separately analysed. At day 10 (Fig. 2.3), there was no significant relationship between survival and humidity at any temperature, indicating that the ticks could survive for this time period even when the humidity was low and the temperature was high. At day 20, there was a significant interactive effect between survival and humidity overall (Z44=4.4, P<0.001). While there was no significant relationship between survival and humidity at 4°C and 15°C, there was a significant relationship at 30°C (Z14=4.6, P<0.001). At day

30, there was again a significant interactive effect between survival and humidity overall (Z44=4.7, P<0.001; Fig. 2.5) and again this was caused by the fact that there was no significant relationship between survival and humidity at 4°C, but there was a significant relationship between survival and humidity at 15°C (Z14=3.7, P<0.001) and at 30°C (Z14=4.2, P<0.001). The relationships between survival and humidity at 4°C, 15°C and 30°C were all significant at day 50 (Z14=3.6, P<0.001; Z14=3.8,

P<0.001; Z14=3.6, P<0.001, respectively; Fig. 2.6) and day 63 (Z14=3.9, P<0.001; Z14=3.6, P<0.001;

Z14=3.1, P<0.05 respectively, Fig. 2.7). The average number of ticks alive was calculated using the data from each of the three repeats and plotted for temperature and humidity at each time point.

30° 3

2 15° Average number alive number Average 1 4°

0 0 20 40 60 80 100

Humidity (% r.h.)

Fig. 2.3 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and 30°) conditions on day 10. The error bars represent standard error of the mean.

3 30° 2 15° 1 4° 0 Average number alive number Average Linear -1 (30°)

-2 0 20 40 60 80 100

Humidity (% r.h.)

Fig. 2.4 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°,

15° and 30°) conditions on day 20 (30°C: Y=0.0787*X - 2.645, Z14=4.6, P<0.001, r²=0.7479). The error bars represent standard error of the mean.

3 30 °C 2 15 °C 1 4°

0 Average number alive number Average Linear (30 °C) -1 Linear (15 °C) -2 0 20 40 60 80 100

Humidity (% r.h.)

Fig. 2.5 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and

30°) conditions on day 30(30°C: Y=0.0619*X - 2.4079, Z14=4.2, P<0.001, r²=0.7261; 15°C: Y=

0.0538*X - 1.2394, Z14=3.7, P<0.001, r²= 0.6418). The error bars represent standard error of the mean.

3 30 °C 2 15 °C

1 4°

Linear

0 (30 °C) Average number alive number Average Linear -1 (15 °C) Linear (4°) -2 0 20 40 60 80 100

Humidity (% r.h.)

Fig. 2.6 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and

30°) conditions on day 50 (30°C: Y=0.0486*X - 1.6685, Z14=3.6, P<0.001, r²=0.6685; 15°C:

Y=0.0534*X - 1.4843, Z14=3.8, P<0.001, r²= 0.6544; 4°C:Y=0.0622*X - 0.7251, Z14=3.6, P<0.001, r²=0.6745). The error bars represent standard error of the mean.

30 °C 2 15 °C 1 4°

0 Linear

Average number alive number Average (30 °C) Linear -1 (15 °C) Linear -2 (4°) 0 20 40 60 80 100

Humidity (% r.h.)

Fig. 2.7 The average number of ticks alive, calculated using the number of ticks alive from each of the three repeats, in different humidity (20, 40, 60, 80 and 95% RH) and temperature (4°, 15° and

30°) conditions on day 63 (30°C: Y=0.0395*X - 1.3973, Z14=3.1, P<0.05, r²=0.5478; 15°C:

Y=0.0486*X - 1.3352, Z14=3.6, P<0.001, r²= 0.4606; 4°C:Y=0.0621*X - 1.1452, Z14=3.9, P<0.001, r²=0.6856). The error bars represent standard error of the mean.

The data were further explored by investigating the influence of saturation deficit on survival rate using a Poisson regression GLM for each time point. At day 10 there was no significant influence (Fig. 2.8). Saturation deficit was found to have a significant logarithmic relationship with survival rate for the remaining time points, at day 20 (Z44=5.6, P<0.001; Fig. 2.9), day 30 (Z44=6.1,

P<0.001; Fig. 2.10), day 50 (Z44=5.8, P<0.001; Fig. 2.11), and day 63 (Z44=5.4, P<0.001; Fig. 2.12).

2 Average Average number alive